ABSTRACT There is general agreement that muscles are utilized by the motor system in a manner that satisfies the mechanical demands of movement, i.e. task dependency. There is also general consensus that muscles are selected by the central nervous system through the integration of centrally generated commands, e.g. CPGs, and various forms of sensory feedback. The structure and function of the CPG and the extent to which various forms of feedback are integrated to yield a task specific response, however, remains open to debate. Furthermore, whether disruption of this feedback after peripheral nerve injury can be compensated remains an issue requiring further study. Immediately after muscle self-reinnervation, we have found clear signs of shortterm compensation for the functional loss of self-reinnervated muscles. Most of these compensatory changes however, disappeared between eight to twelve weeks post trauma as muscle activity patterns, almost all joint kinematic parameters and the joint kinetics of walking became seemingly normal. In the proposed project, we have three major objectives, (1) to understand the mechanical response of the local musculoskeletal system to peripheral nerve injury and repair, (2) to understand the normal and "augmented" progression of motor reconstruction and short-term compensation of muscle coordination during the recovery from self-reinnervation of selected ankle extensors, and (3) to examine the contribution of paw pad cutaneous afferents and group la and Ib afferents from an intact synergist to the adaptation of the motor patterns to the loss of length and possibly force feedback in selected ankle extensors. The intervention to accelerate recovery after selfreinnervation will include the use of an axon growth promoting strategy, i.e. electrical stimulation of nerve branches. We propose to use a self-reinnervated lateral gastrocnemius (LG) and soleus (SOL) model, slope walking to vary environmental demands, and evaluation of joint moments, individual muscle forces, muscle fascicle lengths, muscle reflexes and EMG activity. Afferent feedback will be manipulated using local anasthetic and stimulations of a muscle nerve using implanted nerve cuff in walking cats. We will collaborate very closely with Projects I (English) and III (Nichols) in the collection and interpretation of results, i.e. we will share animal models, resources and expertise with these two other projects.